Autoregressive Moving Average and Generalized Autoregresive Moving Average in Covid-19 Confirmed Cases in Indonesia

TL;DR

This paper applies ARMA and GARMA models to forecast Covid-19 confirmed cases in Indonesia, aiming to identify the most accurate model based on AIC criteria.

Contribution

It evaluates ARMA and GARMA models for Covid-19 case prediction in Indonesia, selecting the best model using AIC for improved forecasting accuracy.

Findings

GARMA model outperforms ARMA in accuracy

Best model identified by lowest AIC value

Model results aid in pandemic data prediction

Abstract

Autoregressive moving average and generalized autoregressive moving average are often used in statistical modeling. This study uses this method because the method uses data from the previous period to model the data for the current period. In addition, the technique is often used in data prediction. The familiar data used is count data. Count data is the data that most often cause data not to spread usually. Therefore, time series modeling, one of which is through arithmetic series, was developed. This study aims to obtain the best modeling results from positive confirmed cases of Covid 19 in Indonesia. They were getting the results from the best modeling for positive confirmed cases of Covid 19 in Indonesia based on the smallest Aikake information criterion value.